Sarah Wahyuni

Something I've been thinking about while building in the small-batch production space (bakeries, candle and soap makers, caterers): almost every tool in this category, including most of what I looked at before building my own, treats "inventory" and "capacity" as the same problem. They're not.

Knowing you have 60 eggs, 8kg of flour, and 1.5kg of butter is an inventory fact. Knowing that translates to 15 cakes, not the 25 a customer just asked for, and that butter is the actual constraint, not flour, is a capacity fact. Most small producers are doing that second translation in their head, every single order, and it's the part that actually determines whether they can say yes.

I ended up building Makeable to handle that translation specifically, rather than trying to be a full inventory or accounting system, because I kept seeing people abandon heavier tools the moment they had to maintain a chart of accounts just to answer a yes/no question (usemakeable.com, for anyone curious what it looks like in practice).

Curious how others here think about this distinction in other domains. Do you see the same "we track the data but not the decision" gap outside of physical production? I'd guess it shows up in capacity planning for services and even dev team throughput, not just materials.

It started with a mess... and almost a scam.

Our founder's personal document storage had turned into complete chaos. He had hundreds of PDFs with no real system, impossible to find anything when it actually mattered. Trying to fix one file, he used a random online PDF editor... which turned out to be a scam tool.

That was the moment DocuAssist.AI was born. A web-based workspace (no app to install) where AI actually helps organize your documents and finds the one you need instantly instead of you scrolling through folders hoping you remember the filename.

Now that we have some distance from the initial AI boom, can you say whether AI took your job or improved it?

I definitely felt the impact in the first half of 2024.